CN108108129B - Method for dynamically quantizing data reliability of solid state disk and solid state disk - Google Patents
Method for dynamically quantizing data reliability of solid state disk and solid state disk Download PDFInfo
- Publication number
- CN108108129B CN108108129B CN201711370601.1A CN201711370601A CN108108129B CN 108108129 B CN108108129 B CN 108108129B CN 201711370601 A CN201711370601 A CN 201711370601A CN 108108129 B CN108108129 B CN 108108129B
- Authority
- CN
- China
- Prior art keywords
- data
- level
- solid state
- state disk
- judgment threshold
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0602—Interfaces specially adapted for storage systems specifically adapted to achieve a particular effect
- G06F3/0614—Improving the reliability of storage systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0628—Interfaces specially adapted for storage systems making use of a particular technique
- G06F3/0638—Organizing or formatting or addressing of data
- G06F3/064—Management of blocks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0628—Interfaces specially adapted for storage systems making use of a particular technique
- G06F3/0646—Horizontal data movement in storage systems, i.e. moving data in between storage devices or systems
- G06F3/0647—Migration mechanisms
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0668—Interfaces specially adapted for storage systems adopting a particular infrastructure
- G06F3/0671—In-line storage system
- G06F3/0673—Single storage device
- G06F3/0674—Disk device
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Digital Magnetic Recording (AREA)
- Debugging And Monitoring (AREA)
- Techniques For Improving Reliability Of Storages (AREA)
Abstract
The invention discloses a method for dynamically quantifying data reliability of a solid state disk and the solid state disk, which are characterized in that a solid state disk system dynamically adjusts a reliability judgment threshold according to the erasing times and the running time of a data block, specifically, the erasing times and the running time are divided into different intervals, each interval is provided with different reliability judgment thresholds, and when the erasing times and the running time of the data block fall into the corresponding interval, the reliability judgment threshold of the data block is dynamically adjusted to the reliability judgment threshold corresponding to the interval. By the method for dynamically quantizing the data reliability judgment threshold of the solid state disk, the influence on the system performance caused by a large amount of data moving actions in a period of time when the system data is kept initialized can be avoided, the data moving triggered for improving the data reliability can be dispersed to different time stages, and the performance of the system can be effectively ensured while the data reliability is considered.
Description
Technical Field
The invention relates to a solid state disk control technology, in particular to a method for dynamically quantizing the data reliability of a solid state disk and the solid state disk.
Background
The degree CECCLevel of the occurrence of CECC is a characteristic quantity representing a reliability judgment threshold, the lower the CECCLevel is, the lower the data is, the corresponding reregrylevel is, and the reliability judgment threshold is represented by the reregrylevel, that is, the reliability judgment threshold is actually the ReadRetryLevel of a certain level, so that the reliability and the reliability judgment threshold are in a negative correlation relationship.
In the design of a firmware system, reliability identification needs to be carried out on data at a NandFlash end, and the reliability of the data is generally classified into Pass, CECC and UECC.
The tagging of data states has several possibilities:
1, normally reading successfully NormalReadSuccess, not triggering rereadReadRetry, and marking the state as Pass;
2 normal reading failure NormalReadFail, triggering rereadretry, and correctly reading data when rereading level ReadRetryLevel < CECClevel, and marking the state as Pass;
3, normal reading failure NormalReadFail, triggering ReadRetry, and correctly reading the data when the ReadRetry Level is more than or equal to CECClevel, wherein the state mark is CECC;
4NormalReadFail, ReadRetryFail, state labeled UECC;
wherein the reliability of the data Pass > CECC > UECC. When data with CECC type reliability appears in firmware design, a moving action is triggered to enhance the reliability of the data, namely, the data is rewritten into the Nand Flash, and errors generated by the internal storage of the Nand Flash are reduced or even eliminated. The UECC type status indicates that the data has not been recoverable.
However, with the use of the solid state disk, the unreliability of NandFlash data is aggravated by the increase of the erasing times PECycle of NandFlash and the increase of the data Retention time. The unreliable result is UECC generated by NormalRead, which needs to rely on ReadRetry function to restore data, and the unreliable aggravation is the sharp increase of ReadRetry level when data is read correctly.
In practical applications, the reliability determination threshold (CECCLevel) is usually a static preset value or a simplified variable value (for example, slcbllock and TLCBlock use different cecclevels). In the background of the prior art, simple selection of a reliability determination threshold can lead to two drawbacks:
the CECClevel independent of PECycle and Retention factors makes the system unable to distinguish the data reliability of the solid state disk in different life stages. In fact, as the PECycle increases and the Retention time increases, the closer the solid state disk is to the end of life, the lower the reliability of data is, and the greater the risk that the data cannot be recovered, so that the safer the solid state disk is, the closer the solid state disk is to the end of life, the lower the CECCLevel value selected by the solid state disk should be, and the reliability of the data should be relatively enhanced through continuous data moving.
In the time period when the solid state disk just keeps operating, the response of the system to the host end is seriously influenced by the frequent data movement, especially the solid state disk at the end of life. The closer to the end of life, the lower the data reliability, the higher the possibility that ReadRetryLevel of data recovery exceeds CECCLevel, so that the possibility and magnitude of data migration are higher. The result is that a large number of data movement actions are focused on the initial period of system data retention. The abstract description is that the reliability of a large amount of data is found to be lower than the reliability judgment threshold value in the current system when the system is just operated and the reliability of the data is evaluated, so that the refreshing of the large amount of data is triggered, and the processing of the system to the request of the Host end is delayed and even blocked.
Disclosure of Invention
Aiming at the defects, the invention aims to dynamically adjust the period data reliability judgment threshold according to different life cycles of the hard disk, so as to reduce the negative influence on the performance while ensuring the data reliability.
The invention provides a method for dynamically quantizing the reliability of solid state disk data, which is characterized in that a solid state disk system dynamically adjusts a reliability judgment threshold according to the erasing times and the running time of a data block, specifically, the erasing times and the running time are divided into different intervals, each interval is provided with a different reliability judgment threshold, and when the erasing times and the running time of the data block fall into the corresponding interval, the reliability judgment threshold of the data block is dynamically adjusted to the reliability judgment threshold corresponding to the interval.
The method for dynamically quantifying the data reliability of the solid state disk is characterized in that the solid state disk is selected by sampling to acquire data, and the reliability judgment threshold CECCLevel corresponding to each interval is acquired, specifically, the selected solid state disk is subjected to an abrasion test, the erasing times PECycle, the data Retention time Retention and the rereadedRetryLevel are recorded, the approximate relation among the erasing times, the data Retention time and the rereading level is acquired, and the reliability judgment threshold of each interval is selected according to the erasing times, the data Retention time and the rereading level corresponding to the interval corresponding to the rereading level.
The method for dynamically quantifying the reliability of the data of the solid state disk is characterized in that the CECClevel at the initial moment is L + RR, the final stable CECClevel is X-RR, after the system data is maintained, the CECClevel is dynamically changed along with the running time, the CECClevel belongs to [ X-RR, L + RR ], along with the increase of the running time of the system, the CECClevel selected by the current system is gradually reduced and approaches to X-RR, and the RR obtains the experience rereading level of each life cycle through tests.
A solid state disk is characterized in that a solid state disk controller dynamically adjusts a reliability judgment threshold value according to the erasing times and the running time of data blocks of each solid state disk of the solid state disk, specifically, the erasing times and the running time are divided into different intervals, each interval is provided with a different reliability judgment threshold value, and when the erasing times and the running time of the data blocks fall into the corresponding interval, the reliability judgment threshold value of the data blocks is dynamically adjusted to the reliability judgment threshold value corresponding to the interval.
The solid state disk is characterized in that the solid state disk is selected in a sampling mode to acquire data acquisition, and the reliability judgment threshold CECClevel corresponding to each interval is acquired.
The solid state disk is characterized in that the CECClevel at the initial moment is L + RR, the final stable CECClevel is X-RR, after system data are maintained, the CECClevel is dynamically changed along with the running time, the CECClevel belongs to [ X-RR, L + RR ], the CECClevel selected by the current system is gradually reduced and approaches to X-RR along with the increase of the running time of the system, and the RR obtains experience re-reading level of each life cycle through tests.
The invention has the beneficial effects that: by the method for dynamically quantizing the data reliability judgment threshold of the solid state disk, the influence on the system performance caused by a large amount of data moving actions in a period of time when the system data is kept initialized can be avoided, the data moving triggered for improving the data reliability can be dispersed to different time stages, and the performance can be effectively ensured while the data reliability is considered.
Drawings
Fig. 1 is a schematic diagram of a SSD mapping management policy in common use at present.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Aiming at the defects of the existing data reliability judgment scheme, the solution of the patent is to dynamically quantify reliability judgment, a reliability judgment threshold value and PECycycle and Retention factors influencing the reliability form a relational expression, meanwhile, a certain relation is formed between the system running time and the reliability judgment threshold value, the reliability judgment threshold value is dynamically set through the factor of the running time of disk data, the data with reliability concentrated in a certain section interval is segmented through different reliability judgment threshold values, further, batch data refreshing is dispersed to different time stages, and meanwhile, the data with lower data reliability is refreshed sooner. In summary, the reliability dynamic judgment scheme determines the upper and lower limits of the reliability judgment threshold value first, and then dynamically selects the reliability judgment threshold value at the current time within the reliability judgment threshold value interval at the system operation time.
The premise of applying the dynamic quantitative reliability judgment scheme is to abstract the relationship between PECycle, Retention and ReadRetryLevel for correctly reading data. The data of the PE Cycle, the Retention and the ReadRetryLevel can be collected in advance in the particle verification stage, and the data are analyzed to derive an approximate relational expression. The PECycle and the Retention values are divided into a certain granularity, and because ReadRetryLevel is insensitive to the change of the PECycle and the Retention values, the PECycle can be divided into 100 granularities, and the Retention can be divided into 1 month granularities, and then the granularity and the ReadRetryLevel form a three-dimensional relation table. The method comprises the steps that a PECycle value and approximate Retention time can be obtained during operation of a firmware system, so that approximate ReadRetryLevel is calculated, and a proper reliability judgment threshold CECCLevel at the initial operation time of the system and a reliability judgment threshold CECCLevel at the final stable operation time of the system are selected according to the ReadRetryLevel value; the CECCLevel at the initial time is larger than ReadRetryLevel, a proper redundancy value + ReadRetryLevel can be selected, the redundancy value is L, the selected CECCLevel can ensure that the proportion marked as a CECC state at the initial time is as low as possible, the L value cannot be larger, and otherwise, the reliability at the initial time cannot be ensured.
Fig. 1 is a schematic flow chart of dynamic adjustment, where whether CECCLevel needs to be updated is determined by whether current CECCLevel reaches final CECCLevel and whether current system running time reaches an update time point. And (3) recording the Read RetryLevel deduced from the relation table of PECycle, Retention and ReadRetryLeve as RR, wherein the CECCLevel at the initial moment is L + RR, and the stable CECCLevel is X-RR finally. After the system data is kept, selecting CECClevel at the initial moment, wherein the CECClevel dynamically changes along with the operation time, the CECClevel belongs to [ X-RR, L + RR ], and the CECClevel selected by the current system gradually decreases and approaches to X-RR along with the increase of the operation time of the system. For the disks in different life stages, the RR is different, and the reliability judgment threshold value selection interval is also different. The smaller the PERCYCLE is, the shorter the Retention is, the higher the data reliability is, and the smaller the RR is, the smaller the reliability judgment threshold interval range is (the interval range is L-X +2 RR).
While the invention has been described with reference to a particular embodiment, it will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention as defined by the appended claims.
Claims (4)
1. A method for dynamically quantifying data reliability of a solid state disk is characterized in that a solid state disk system dynamically adjusts a reliability judgment threshold according to the erasing times and the running time of a data block, specifically, the erasing times and the running time are divided into different intervals, each interval is provided with a different reliability judgment threshold, and when the erasing times and the running time of the data block fall into the corresponding interval, the reliability judgment threshold of the data block is dynamically adjusted to the reliability judgment threshold corresponding to the interval; the method comprises the steps of acquiring data through sampling and selecting a solid state disk to acquire a reliability judgment threshold CECC Level corresponding to each interval, specifically, performing an abrasion test on the selected solid state disk, recording erasing times PE Cycle, data Retention time Retention and rereading Level Read Retry Level, acquiring an approximate relation among the erasing times, the data Retention time and the rereading Level, and selecting the reliability judgment threshold of each interval according to the erasing times, the data Retention time and the rereading Level corresponding to the interval corresponding to the rereading Level.
2. The method for dynamically quantifying reliability of data in a solid state disk according to claim 1, wherein the CECC Level at the initial moment is L + RR, the final stable CECC Level is X-RR, after the system data is maintained, the CECC Level dynamically changes along with the running time, the CECC Level belongs to [ X-RR, L + RR ], along with the increase of the running time of the system, the CECC Level selected by the current system gradually decreases and approaches to X-RR, and the RR obtains the empirical rereading Level of each life cycle through tests.
3. A solid state disk is characterized in that a solid state disk controller dynamically adjusts a reliability judgment threshold value according to the erasing times and the running time of data blocks of each solid state disk of the solid state disk, specifically, the erasing times and the running time are divided into different intervals, each interval is provided with a different reliability judgment threshold value, and when the erasing times and the running time of the data blocks fall into the corresponding interval, the reliability judgment threshold value of the data blocks is dynamically adjusted to the reliability judgment threshold value corresponding to the interval; the method comprises the steps of acquiring data through sampling and selecting a solid state disk to acquire a reliability judgment threshold CECC Level corresponding to each interval, specifically, performing an abrasion test on the selected solid state disk, recording erasing times PE Cycle, data Retention time Retention and rereading Level Read Retry Level, acquiring an approximate relation among the erasing times, the data Retention time and the rereading Level, and selecting the reliability judgment threshold of each interval according to the erasing times, the data Retention time and the rereading Level corresponding to the interval corresponding to the rereading Level.
4. The solid state disk of claim 3, wherein the CECC Level at the initial time is L + RR, the final stable CECC Level is X-RR, after the system data is maintained, the CECC Level is dynamically changed with the running time, the CECC Level belongs to [ X-RR, L + RR ], and as the running time of the system increases, the CECC Level selected by the current system gradually decreases and approaches X-RR, and RR obtains the experience re-reading Level of each life cycle through tests.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711370601.1A CN108108129B (en) | 2017-12-19 | 2017-12-19 | Method for dynamically quantizing data reliability of solid state disk and solid state disk |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711370601.1A CN108108129B (en) | 2017-12-19 | 2017-12-19 | Method for dynamically quantizing data reliability of solid state disk and solid state disk |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108108129A CN108108129A (en) | 2018-06-01 |
| CN108108129B true CN108108129B (en) | 2020-10-20 |
Family
ID=62211023
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711370601.1A Active CN108108129B (en) | 2017-12-19 | 2017-12-19 | Method for dynamically quantizing data reliability of solid state disk and solid state disk |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108108129B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109460191A (en) * | 2018-11-15 | 2019-03-12 | 苏州韦科韬信息技术有限公司 | A kind of method of dynamic quantization solid state hard disk data reliability |
| CN110109714B (en) * | 2019-04-25 | 2022-03-29 | 深圳忆联信息系统有限公司 | Method and device for improving firmware loading efficiency, computer equipment and storage medium |
| CN115904256B (en) * | 2023-02-20 | 2023-06-13 | 中电装备山东电子有限公司 | Data storage method and memory |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106484761A (en) * | 2015-08-24 | 2017-03-08 | Hgst荷兰公司 | For improving the method and system of storage daily record |
| US9798656B2 (en) * | 2013-06-28 | 2017-10-24 | Samsung Electronics Co., Ltd. | Memory controller, method of operating, and apparatus including same |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR102456104B1 (en) * | 2015-08-24 | 2022-10-19 | 삼성전자주식회사 | Method for operating storage device changing operation condition depending on data reliability |
-
2017
- 2017-12-19 CN CN201711370601.1A patent/CN108108129B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9798656B2 (en) * | 2013-06-28 | 2017-10-24 | Samsung Electronics Co., Ltd. | Memory controller, method of operating, and apparatus including same |
| CN106484761A (en) * | 2015-08-24 | 2017-03-08 | Hgst荷兰公司 | For improving the method and system of storage daily record |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108108129A (en) | 2018-06-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106776095B (en) | Intelligent detection method and detection device for data reliability of SSD (solid State disk) | |
| CN108108129B (en) | Method for dynamically quantizing data reliability of solid state disk and solid state disk | |
| US11392328B2 (en) | Dynamic background scan optimization in a memory sub-system | |
| KR101348665B1 (en) | Method for estimating and reporting the life expectancy of flash-disk memory | |
| CN108733579B (en) | Method for providing window of opportunity hints for background operations in solid state disk and solid state disk | |
| CN111090398A (en) | Garbage recycling method, device and equipment for solid state disk and readable storage medium | |
| CN108646977B (en) | SSD (solid State disk) cold and hot data judgment method and garbage recycling method | |
| CN110347335B (en) | Solid state disk data storage method and device | |
| US11587628B2 (en) | Refreshing data stored at a memory component based on a memory component characteristic component | |
| CN112817527B (en) | Power-down data storage method, singlechip and computer-readable storage medium | |
| US11688467B2 (en) | Defect detection in memories with time-varying bit error rate | |
| US20210295900A1 (en) | Performing a refresh operation based on a characteristic of a memory sub-system | |
| CN111400204A (en) | Solid-state disk caching method, system and related equipment | |
| CN107832202A (en) | A kind of method, apparatus and computer-readable recording medium for detecting hard disk | |
| CN110764960B (en) | Solid state disk firmware testing method | |
| US20230297263A1 (en) | Method and system for testing conversion relationship between block reading and page reading in flash memory chip | |
| US11404141B2 (en) | Preemptive read refresh in memories with time-varying error rates | |
| CN112732181A (en) | Data migration method of SSD and related device | |
| CN109669828B (en) | Hard disk detection method and device | |
| WO2020050953A1 (en) | Accessible accumulated memory temperature readings in a memory sub-system | |
| CN109460191A (en) | A kind of method of dynamic quantization solid state hard disk data reliability | |
| CN112732179A (en) | Data management method of SSD (solid State disk) and related device | |
| CN114495996B (en) | Method and device for controlling power consumption of solid state disk, computer equipment and medium | |
| CN117275561B (en) | Method and system for adjusting wear balance force of solid state disk | |
| CN113918094B (en) | SSD writing bandwidth control method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |